It has been determined that a set of genes present in various strains of retroviruses is respon- sible for tumors in animals. Sequences homologous to these oncogenes can be found in normal cells of most animals from invertebrates up to and including man. Due to their highly conserved sequences and wide distribution, it has been suggested that these genes play some essential role in the maintenance of cells and that an alteration either in the sequence or the level of expression may cause abnormal cell growth which would result in tumor formation. As our overall objectives, we are in the process of determining which, if any, of these genes may be involved in human leukemias and what functions these highly conserved genes have in normal cells. We will determine whether the level of expression of several known hematopoietic-associated oncogenes in highly characterized childhood leukemia cells is abnormal as compared to levels in normal hematopoietic cells at the same stage of differentiation. To accomplish this, we will employ in situ hybridization techniques to observe oncogene RNA levels in morphologically identifiable cells. In addition, we have isolated a human myc-related sequence from a library containing A673 rhabdomyosarcoma cell DNA. This cloned sequence has 5' and 3' c-myc homology but is substantially different than previously isolated c-myc and N-myc clones. Upon transfection the cloned DNA will transform primary mouse macrophages, some of which are tumorigenic in nude mice. We plan to determine what portion of the cloned DNA has the biological activity using transfection of subcloned fragments and to determine the sequence of this gene for comparison with other human myc-related sequences. (M)